View metadata, citation and similar papers at core.ac.uk brought to you by CORE Published in $QGURORJ\ ± which should be cited to refer to this work. provided by RERO DOC Digital Library Correspondence: Charles Sultan, Unite d’Endocrinologie- Endocrine and molecular Gynecologie Pediatriques, Departement de Pediatrie, Hopital^ Arnaud-de-Villeneuve, CHU investigations in a cohort of 25 Montpellier et Universite Montpellier 1, Montpellier, France adolescent males with prominent/ E-mail: [email protected] persistent pubertal gynecomastia Keywords: adolescence, disorders of sex differentiation, sex 1,2F. Paris, 1,3L. Gaspari, 4F. Mbou, 2P. Philibert, 2F. Audran, 5Y. Morel, hormones, steroids, gynecomastia 6A. Biason-Lauber and 1,2C. Sultan 1Unite d’Endocrinologie-Gynecologie Pediatriques, Departement de Pediatrie, Hopital^ Arnaud-de- Villeneuve, CHU Montpellier et Universite Montpellier 1, Montpellier, France, 2Departement d’Hormonologie (Developpement et Reproduction), Hopital^ Lapeyronie, CHU Montpellier et Universite Montpellier 1, Montpellier, France, 3Departement de Pediatrie, Hopital^ Caremeau, CHU N^ımes, N^ımes, France, 4Departement de Pediatrie, CHU de Fort de France, Martinique, 5Centre de Biologie et Pathologie Est, Bron, France, and 6Department of Medicine, University of Fribourg, Fribourg, Switzerland SUMMARY Pubertal gynecomastia is a common condition observed in up to 65% of adolescent males. It is usually idiopathic and tends to regress within 1–2 years. In this descriptive cross-sectional study, we investigated 25 adolescent males with prominent (>B3) and/or persistent (>2 years) pubertal gynecomastia (P/PPG) to determine whether a hormonal/genetic defect might underline this condi- tion. Endocrine investigation revealed the absence of hormonal disturbance for 18 boys (72%). Three patients presented Klinefelter syndrome and three a partial androgen insensitivity syndrome (PAIS) as a result of p.Ala646Asp and p.Ala45Gly mutations of the androgen receptor gene. The last patient showed a 17a-hydroxylase/17,20-lyase deficiency as a result of a compound heterozygous mutation of the CYP17A1 gene leading to p.Pro35Thr(P35T) and p.Arg239Stop(R239X) in the P450c17 protein. Enzymatic activity was analyzed: the mutant protein bearing the premature stop codon R239X showed a complete loss of 17a-hydroxylase and 17,20-lyase activity. The mutant P35T seemed to retain 15–20% of 17a-hydroxylase and about 8–10% of 17,20-lyase activity. This work demon- strates that P/PPG had an endocrine/genetic cause in 28% of our cases. PAIS may be expressed only by isolated gynecomastia as well as by 17a-hydroxylase/17,20-lyase deficiency. Isolated P/PPG is not always a ‘physiological’ condition and should thus be investi- gated through adequate endocrine and genetic investigations, even though larger studies are needed to better determine the real prevalence of genetic defects in such patients. http://doc.rero.ch INTRODUCTION In prominent and/or persistent pubertal gynecomastia (P/ Pubertal gynecomastia is a common condition observed in PPG), the imbalance between the actions of androgens and up to 65% of adolescent males. It results from a proliferation estrogens on male breast tissue may be attributable to an excess in the glandular component of the breast, usually because a of estrogen production, a decrease in androgen production or transient relative imbalance between androgens and estrogens. action, or drugs. Estrogen-producing gonadal or adrenal tumors Pubertal gynecomastia is generally idiopathic and the breast and endocrine disorders thus have to be looked for, and patients development is most often Tanner stage B2 (Tanner, 1962). should be carefully questioned about drug use. Moreover, aro- As it tends to regress within 1–2 years, pubertal gynecomastia matase excess has recently been reported as a cause of P/PPG is considered as a benign condition and is regarded as a (Fukami et al., 2011). Careful attention should be paid to the part of normal development in pubertal males (Moore et al., patient’s morphotype and the testes should be examined to 1984). exclude Klinefelter syndrome. An underlying disorder of sex dif- If male breast development is over B3-B4, however, or if it per- ferentiation (DSD) such as partial androgen insensitivity syn- sists beyond 2–3 years, it may be the sign of an endocrine disor- drome (PAIS) or 17-b-hydroxysteroid dehydrogenase defect der, becoming a source of anxiety, self-consciousness, should also be ruled out (Narula & Carlson, 2007). In addition, embarrassment and severe psychological discomfort. In these environmental factors (Ma & Geffner, 2008) may be implicated. cases, determining the pathogenesis of pubertal gynecomastia Any of these conditions can cause excessive estrogen and/or becomes mandatory. deficient androgen activity. 1 The aim of our study was to identify the causes of prominent/ 17-OHP: 2.7 and 2.9%, DHEA: 6.8 and 7.9%, Δ4: 5.6 and 6%, pro- persistent pubertal gynecomastia (P/PPG) in adolescent males, gesterone: 2.5 and 3.5%, LH: 4.8 and 6.6%, FSH: 4.9 and 3.5, PRL: particularly hormonal/genetic defects. 2.1 and 2.9%, TSH: 3 and 6%, and cortisol: 2.6 and 3.8%. The detection limits were as follows: E2: 5 pg/mL, T: 0.04 ng/mL, 17-OHP: 0.1 ng/mL, DHEA: 0.3 ng/mL, Δ4: 0.04 ng/mL, proges- SUBJECTS AND METHODS terone: 0.09 nmol/L, LH: 0.1 U/L, FSH: 0.1 U/L, prolactin: Patients 0.24 ng/mL, TSH: 0.005 lU/mL, and cortisol: 0.04 lg/dL. This descriptive cross-sectional study included all adolescent males referred to the Pediatric Endocrine Clinic of the Montpel- Karyotyping lier University Hospital between January 2011 and January 2014 Karyotyping was performed when the clinical and biological for prominent and/or persistent pubertal gynecomastia. The investigations suggested a genetic disorder; i.e., when the basal gynecomastia was evaluated by inspection and palpation as FSH level was very high and/or the testicular volume was small. described by (Braunstein, 2007) to distinguish between pseudog- It was also performed for a patient in whom we suspected a ynecomastia and true gynecomastia. The selection criteria steroidogenesis defect. included the following: (i) Tanner pubertal status ≥2(Tanner, 1962), (ii) lack of associated genital abnormalities, (iii) prominent Molecular analysis breast development (>B3), or (iv) persistence beyond 2 years. All patients were asked about the use of drugs known to be involved Androgen receptor (AR) gene in the development of gynecomastia, i.e., spironolactone, cime- After polymerase chain reaction (PCR) amplification of exons tidine, ketoconazole, and antiandrogens (Deepinder & Braun- 1–8 of the AR using the Taq PCR Master Mix kit from Qiagen stein, 2012), as well as the abuse of alcohol, marijuana, heroin, (Courtaboeuf, France), we performed direct sequencing using amphetamines, and anabolic steroids (Goldman, 2010). Renal the BigDye terminator v1.1 kit (Applied Biosystems, Foster City, and liver diseases were excluded for all patients. CA, USA) and an ABI Prism310 Genetic Analyzer (Applera, Cour- Approval for this study was first obtained from the appropriate taboeuf, France), as reported elsewhere (Philibert et al., 2010). In institutional review board and all patients gave signed informed cases of mutation, PCR and sequencing of the DNA sample were consent. repeated twice to confirm the finding and rule out any PCR-gen- erated errors. Every PCR product was sequenced with forward Anthropometric data and reverse primers. The amino acid numbering for the AR was Standing height was measured to the nearest 0.1 cm with a based on the NCBI reference sequence NM_000044.2 and the AR stadiometer (Seca, Semur-en-Auxois, France). Weight was mea- database (Gottlieb et al., 2012). sured on a weight scale with a precision of 0.1 kg. Body mass index (BMI) was calculated as weight (kg) divided by the square CYP17A1 gene of height (m2). Grades of obesity were defined on the basis of the The CYP17A1 gene was studied when a P450c17 enzymatic cut-off points used by the Obesity Task Force and derived from deficiency was suspected on the basis of the patient’s hormone Cole’s work (Cole et al., 2000). Pubertal development was profile. After obtaining informed consent, genomic DNA was assessed by breast stages 1–5 of the Tanner classification (Tan- extracted from peripheral leukocytes and then used to perform ner, 1962). PCR exonic amplification of the gene, as previously described (Biason-Lauber et al., 1997). The PCR products were sequenced http://doc.rero.ch Radiological investigation using the Big Dye Terminator Cycle Sequencing Kit and analyzed Bone age (BA) was determined using the Greulich and Pyle on an ABI Prism 310 Genetic Analyzer (Applied Biosystems). Ref- method and testicular ultrasonography was performed for all erence Sequence: RefSeq NM_000102.3. Primer sequences are patients to exclude a testicular tumor. available upon request. Concerning the molecular analysis, patient sequences were Hormonal (biological) studies compared with the sequence of a control patient with normal XY The following hormones were measured at baseline for all DNA. patients: estradiol (E2) (bioMerieux, Craponne, France), testos- terone (T), D4-androstenedione (D4) (Immunotech, Marseille, Expression studies France), dehydroepiandrostenedione (DHEA) (Beckman Coulter, Expression studies were performed as previously described Villepinte France), luteinizing hormone (LH), follicle-stimulating (Rosa et al., 2010). Briefly, wild-type CYP17A1 cDNA (originally hormone (FSH) (bioMerieux, Craponne, France), prolactin from Michael R. Waterman) was inserted into a pcDNA3.1 vector (BRAHMS, Asnieres-sur-Seine, France), and thyroid-stimulating after addition of an N-terminal myc-tag. Mutant cDNAs were hormone (TSH) (Roche, Boulogne-Billancourt, France). constructed using the QuikChange II site-directed mutagenesis The ACTH test was performed in one patient with the mea- kit from Stratagene (La Jolla, CA, USA). Introduction of the surement of cortisol (Roche), 17-OHP (CIS bio international, Gif- mutations was confirmed by sequencing. Wild-type or mutant sur-Yvette, France), progesterone (Roche), and DHEA before and cDNA was transiently transfected into confluent COS-1 cells after stimulation.